Fulmemo is a focused production-engineering publication for people operating web, backend, data, and real-time systems. The guides are written as diagnostic runbooks: establish scope, preserve evidence, change one layer, verify the original failure, and keep a rollback point.
Before running a command
- Confirm the site, path, user, and environment named by the command.
- Make a restorable database-and-files backup before state-changing work.
- Prefer staging for plugin, theme, PHP, rewrite, and database tests.
- Do not paste passwords, cookies, tokens, private keys, database exports, or personal information into public support channels.
Choose the guide by symptom
Start with the response or failure you can observe—not the component you suspect. WordPress categories cover runtime, access, routing, integrations, indexing, and performance. Java Services covers concurrency, garbage collection, deadlines, and retries. Go Runtime covers goroutine ownership, scheduler traces, allocation pressure, and garbage collection. Programmatic Advertising covers OpenRTB outcomes, public supply-chain authorization, and browser auction timing. Distributed Data covers replicas, sharding, slots, and failover. Web Latency covers critical rendering and shared caching. Real-time Games covers prediction, reconciliation, capacity, and load tests.
Read logs safely
Match the failing request to a UTC timestamp and capture the first relevant error. Keep production errors out of page output. Redact secrets and personal data before sharing evidence.
Know when to stop
Stop when the next step requires access you do not have, a destructive database operation, an untested restore, or a security decision outside your role. A concise escalation with evidence is better than an improvised production change.